Predicting Startup Success, a Literature Review
Authors: ['Harjo Baskoro']
Year: 2022
Methodology
- Sample: 132
- Design: literature_review
- Data: Academic journals, Conference proceedings (ICoSTEC), Existing success models (e.g., Nalintippayawong model)
Factors Extracted (7)
Customer Perspective [anecdotal] — Positive effect on market opportunity
Market Opportunity [anecdotal] — Positive effect on business model and support partners
Business Model [anecdotal] — Positive effect on potential startup success
Support Partner [anecdotal] — Positive effect on potential startup success
Potential of a Startup [anecdotal] — Directly related to success (profit/fundraising)
Founding Team Experience [anecdotal] — Not specified (part of 132 identified variables)
Financial Resources/Fundraising [anecdotal] — Used as a success metric
Key Findings
- The startup failure rate is approximately 90%, necessitating the use of predictive models to identify the successful 10%.
- A systematic review identified 132 distinct factors influencing startup success across various literature sources.
- Success is most effectively defined and measured by a combination of operational profit and the ability to raise funds.
Limitations
- The high number of variables (132) creates a 'curse of dimensionality' for machine learning models, requiring significant feature reduction.
- The study notes high uncertainty in the initial stages of a startup, making early-stage prediction inherently volatile.
- The paper is a literature review and does not present a new primary empirical dataset, relying instead on existing models like Nalintippayawong.
Extracted by lib/ingest/literature_review.py via gemini-flash